Lumens Not Always Equivalent to Wattage?

This is a technical question for experts on bulbs and why some bulbs of the SAME wattage are said to have higher lumen output.

My problem is not about flashlights but my crummy Ford Explorer's headlamp system.

In studying lamps and their luminous output vs. wattage, I am finding specs that don't relate clearly to wattage. It seems that some bulbs of the same wattage have more luminous intensity. The Sylvania SilverStar bulbs, for instance are maybe 8% brighter than OEM bulbs of the same 55-watt rating.
Some 130-watt bulbs I tried were no brighter than the SilverStars.

Now I know my testing was flawed because the stock wiring in the Ford is absurdly under-capacity. There was almost 3 volts dropped in the wiring to the bulbs, referenced to the battery voltage. The higher wattage bulbs got less power for this reason.

But even all things being equal, some bulbs are brighter.
I can think of only two things:

the brighter bulb of a given wattage has a clearer quartz bulb that lets a higher percentage of light pass through.
The brighter bulb probably runs the filament at a higher temperature (shorter filament or fewer turns of tungsten wire?)

There may be other reasons. But I was interested in hearing from our bulb experts on the physics behind luminous intensity and why two 55-watt bulbs are not equal intensity. Is it the reasons I cited, or is there more to it?

How are you comparing the lumen output? If it's by the intensity of the projected beam, that's something that depends as much (sometimes more) upon focusing than upon lumens. Even if you are using the same reflector, a smaller light source will result in a tighter focus.

Same reflector and lens. Some bulbs are a bit whiter, so perhaps the filament voltage rating is slightly lower, but I wonder if there is more to it than that. PIAA makes some very expensive 55W halogen bulbs that they claim have the lumen output of competitor's 100W halogen bulbs, for instance.

I noticed that the original H4 bulb in my Thor spotlight was brighter than another H4 bulb that I got out of a headlamp assembly (both bulbs were 90/100W versions).

I was starting to wonder if there is something about the way the filament is wound (to get more surface area?) or even the transmissivity of the quartz bulb. The cheap Chinese import bulbs seem to have the lower output, while the costly bulbs seem as bright in lower wattage versions.

What I'm trying to establish is whether a $78 pair of PIAAs is going to produce more lumens per watt than a $13 pair from Wal-Mart or a local parts store.

mweiss-

In answer to your post title, lumens and watts are not equivalent. In fact lumens and watts have no real direct relationship in physics. The equating of light and watts is something that we have all learned, because it seemed convenient under certain circumstances. For example, if you want to make the ceiling fixture in your bedroom brighter, how to you do it? Take out the 60W bulb (about 800 lumens) and put in a 75W bulb (about 1100 lumens). Much brighter! However, you don't "see" an increase in watts, but an increase in lumens. We simply needed more power to produce more lumens. We all have learned this relationship between lumens and watts, which as you have discovered does not hold up.

The watt is used to specify the rate at which electrical energy is dissipated, or the rate at which electromagnetic energy is radiated, absorbed, or dissipated.

The definition of a watt does not mention light because there is no direct relationship.

As for your car, you have discovered that there are many more factors than watts in play. As has been mentioned, the quality and condition of the reflector is a huge factor. Also the type of coating on the lamp envelope makes a big difference. Another factor in cars is the outside enclosure. Most are acrylic, and as they age they slowly loses efficiency in transmittance of light.

Also disregard such claims as 8% increase in output. The human eye can't perceive such a small difference. You can perceive a fairly small difference in color temperature which is why many of the aftermarket lamps are pushed into the high white/blue range. Can you see better? Probably not much.

You might gain more improvement, by curing your voltage drop to the lamp if it is excessive.

Mark

I agree with all of your points. I do have an electrical engineering background, but I am not as well versed in halogen lights and whether there are differences possible due to filament winding patterns, bulb material, or other factors.

So from your response, I guess it's safe to conclude that any cheap bulb should have the same brightness if the filament is burning at the same temperature as another bulb it is compared with. So whether I buy the $13.78 pair or the $78 PIAA pair would make no difference if both are clear bulbs with no blue coating.

I made up the 8% figure based on my visual perception of the SilverStars over my stock bulbs. Same wattage. They are whiter and do provide more light. But not 1/10th as much as my wife's Mitsubishi with its stock lamps. I can drive her car with low beams and almost never need the high beam, and when I do, they're almost too bright for my eyes.

It's a tough call. You didn't say the vintage of your Explorer, but I might guess it uses 9000 series lamps. I went through this same exercise a while back with my Grand Cherokee. I ended up with these: Sylvania From Pep Boys, about $16 a pair.
Did I see the dramatic difference they show in the picture? NO! I think skip any that have blue this or that. Just a fashion thing to mimic HID lamps, IMO.

I still believe the biggest limitation is in the design and materials of the housing. Can a pair of $80 bulbs overcome this and make a large improvement? I wasn't willing to bet on it.

Remember, too they are $40 each when they burn out. Also you noticed that your wife's Mitsu is far superior stock. Better design and materials. Headlight output is regulated by DOT and her car is limited by the same rules as yours.

Headlamp design on American cars has always lagged those from Europe and now Asia. Getting better, but still a long way to go.

mweiss said:

So from your response, I guess it's safe to conclude that any cheap bulb should have the same brightness if the filament is burning at the same temperature as another bulb it is compared with. So whether I buy the $13.78 pair or the $78 PIAA pair would make no difference if both are clear bulbs with no blue coating.

Click to expand...

There are other factors that also must be considered.

The output from "long lasting" or heavy duty bulbs when compared to the same wattage will usually be a lot less.

The LightForce GL05: 75 watt, 12 volt light shows 150,000 cp and the GL07: 75 watt, 12 volt light shows 250,000 in the same housing. The 05 is the longlife version.

I'm guessing that they use a thicker wire that's either longer or has more resistance.

short answer:
"Watts' is the amount of electical power consumed by the bulb
"Lumens" is less you he brightest spot in the beam.
There are a lot of variables when you convert from one (power consumption) to the other (light brightness).

A more efficient light emitter may put out more light at less wattage as a conventional emitter (e.g. LED vs. Incandescent). A better shaped reflector may focus more of the output in spotlight beam (more lumens) or spread it out as a floodlight (less lumens).

Given lights with similar desgiin goals you'll probably see some trends emerge, but don't place too much value on a figure like "lumens per watt" unless the related tradeoffs fit your need.

No, lumens do not depend on focus. Lumens are a measure of the total light output.

Yup, my Explorer is a '98 and uses 9007 bulbs.
Running Sylvania SilverStars in there now, which are the brightest 55W bulbs I've tried so far.
I too have concluded that the housing/lens is the problem, although the high beam is limited by not having four lamps burning.
I'm addressing the housing with the diamond cut reflector assemblies, which I bought and am awaiting shipment of.
The next step is the wiring harness, which I ordered from SUVLights.com yesterday.
After both changes, I hope to see a drastic improvement in light pattern and intensity in the areas of importance.

Regarding factors in bulb intensity, I think that the long lasting versions are just higher voltage, longer filaments that burn at lower color temperature. Possibly the SilverStar bulbs I have now may burn out soon, because they are running whiter than my stock bulbs.
If there is no trick to the glass composition, then it's all a matter of filament thickness and length, in which case, I suppose that if I find a pair of 100/130W bulbs rated for 12V instead of 13.8V, I should have pretty white light output with new harness and new reflector housings. Failing that, it's time to go HID.. some guy on the Explorer forum actually hacked up a new set of diamond cut housings and installed BMW projector HIDs in his. His beam patterns look ideal, like an HID version of my old Cibie halogen lamps. But that's a lot of work, risky and big bucks...
I figure the diamond cut reflectors, clear lenses and a new wiring harness ought to bring the lighting into parity with the wife's Mitsubishi.

This may be highjacking this thread a little but it's on the same theme
this sites quotes the overdriver mr16 20watt bulbs as 485lumens (overdriven to 23watts) and the 35watt MR16bulb as 830watts (overdriven to 40watts) is this down to the higherwattage bulb being less efficient or has the company taken the voltage sag into account to come up with the less lumens? (20*1.75 should make the 35watt bulb 41watts when overdriven and 848lumens if it was a linear relationship)

Also, don't take the PIAA claims as gospel. I've used their product in the past, but do not any longer. All measurements that PIAA uses to make the 55W=100W comparisons are flawed, as they only measure with an intensity device at certain locations in the beam. A different focus thereby equals a "higher" wattage? Nonsense.

This is a decent link on your topic: http://dsl.torque.net/home.html

I conversed with Mr. Stern in a couple of emails, and to paraphrase a rhetorical question he posed about PIAA bulbs: "how are they going to explain putting blue/purple Schmutz on the bulb (to "increase" colour temperature by blocking other yellow/red colours), and an *increase* in light output at the same time?" Yes, I'm certain he used the word "schmutz"

I believe that he actually recommended the Sylvania Xtravision bulbs as being a for-real improvement.

-Trev

Thanks for steering me to that web site. Although most of it is stuff I've already read and understood based on my limited optical background with laser optics 30 years ago, the part where he's talking about "+50" high efficiency bulbs and short lifetimes is answering my question. It appears that they burn the filaments hotter and thus whiter light output and more lumens. So that's the type of bulbs I need.
Common sense told me that blue coatings attenuate some of the red/yellow light wavelengths, so are less bright.
I had some Korean import 100/130 bulbs with blue coating that I scraped off with a razor blade. However, the sockets soon melted and I didn't have much more light output anyway.
Since I'm already running the XtraVision bulbs, I guess I need higher wattage and higher voltage as well to get them burning white.

I second the mention in the earlier post stating that the single most critical factor in increasing automotive headlamp performance with aftermarket high output bulbs is decreasing resistance from the mile-long 16ga path the electricity has to travel. Best way I found to do that was use a pair of Bosch 5-pin relays with the stock wiring used as the relay trigger, then 20A fused heavy gauge (usually 12 or even 10) short runs of wire directly to the positive terminal of the battery (fused there, natch) to supply the power. The theory is you use the stock wiring only as a low current switch to enable the high-current (read: very low voltage drop) power to the bulbs. I used to measure more than a 1.5V drop from substituting 8A 100W bulbs in place of 55W bulbs in some instances. This applies for incan bulbs specifically, the newfangled HID retrofits are nicely self-contained from what I've seen. When strangled by the resistance in the long factory circuit path, the higher the intended output of the bulb, the bigger the performance hit / greater voltage drop. Still a net increase in brightness, but NOTHING compared to getting the at-the-bulb voltage as close to alternator voltage as possible.

As to the lumens per watt issue, that generally refers to overall efficiency of the entire system at work to produce the light. Filament physical characteristics, glass envelope shape and properties, gas fill chemical properties and effect on redeposition of tungsten particles on the filament, all these tiny details form a kind of black art in bulb design that ultimately can be expressed in a lumens per watt ratio at a given electrical drive level.

Thats just the light source (incan), which is a mini-universe of variables. The light then radiates out into the bean-counter-ruled domain of reflector design, or just what portion of the vehicle budget did we reserve for the shiny aluminum thingy behind the flimsy plastic lens. Not applicable to many cars nowadays; sarcastically referring to the great "aero" headlight revolution that steered away from easily replacable modular rectangular lights. Think intro of the Ford Taurus. I'm dating myself.

Quality bulbs are made to a high standard and generally are more efficient and dont rely on marketing hype to sell. beware of inexpensive bulbs with gosh-golly-gee-whiz numbers and hype trying for your attention. PIAA IMHO is a well-regarded manufacturer. Philips, Osram, Hella, Narva, Bosch all top drawer. The OEM replacement parts are a snooze, but careful shopping and you can find treasures of hi-po replacements.

AndrewL said:

this sites quotes the overdriver mr16 20watt bulbs as 485lumens (overdriven to 23watts) and the 35watt MR16bulb as 830watts (overdriven to 40watts) is this down to the higherwattage bulb being less efficient or has the company taken the voltage sag into account to come up with the less lumens? (20*1.75 should make the 35watt bulb 41watts when overdriven and 848lumens if it was a linear relationship)

Click to expand...

What the Lumicycle site is talking about is overvolting their lamps. They run 12v MR11s at 13.2v. This apparently gives a 40% increase in light output for ony a 17% increase in power consumption (the figures obviously aren't precise, elsewhere they quote 38% & 15% respectively). It also significantly decreases the life of the bulb, by about 2/3 I gather (but MR11s have a nominal lifespan of 1000 hours to start off with).
Doing the sums using the quoted lumen outputs and new wattages I got very similar figures - about 20 lumens/watt.

I think the part of this that's relevant to mweiss is how the voltage supplied to the lamps can effect the light output. If you're getting a 3v drop in the wiring your car headlamps are only getting 9v (please tell me it isn't a 24v truck system!) - 75% of what they're designed for. Have you ever tried putting 1.2v rechargeable cells in a Maglite with a stock lamp?

Just my 2p worth - but I'd go with the suggestion that improving the wiring is your best bet. Another tip (excuse me if this is a lesson in sucking eggs) is to be really careful about keeping the lamps clean when you install them, it's amazing how yellow a headlight can go when there's a tiny layer of deposited grease (vapourised off the lamp) on the reflector.